Resource management original
Download as PPTX, PDF•0 likes•445 views
The document discusses resource management in distributed systems, focusing on the dynamic allocation of resources by operating systems to competing processes. It explains various scheduling algorithms, including load-balancing and load-sharing approaches, which aim to optimize performance by distributing tasks across nodes and maintaining balanced workloads. Key considerations include static versus dynamic load balancing, communication costs between tasks, and the importance of quick decision-making, fault tolerance, and fairness in service delivery.
Resource management original
- 1. Resources Management & Features of Global Scheduling Algorithm Usman Namadi Inuwa
- 2. ● Resource Management is the dynamic allocation and de-allocation of resources by an OS to processes that compete for those resources. ● Distributed systems contain a set of resources interconnected by a network ● Processes are migrated to fulfill their resource requirements ● Resource manager control the assignment of resources to processes ● Resources can be logical (shared file) or physical (CPU)
- 3. Task Assignment Approach Load-Balancing Approach Load-Sharing Approach
- 4. Each process submitted is viewed as a collection of related tasks. Tasks are scheduled to suitable nodes to improve performance. Assumptions of task assignment: A process has already been split into pieces (task) Amount of comp. and the speed of node is known The cost of processing each task is known The IPC cost between every pair of task is known Precedence relation among the tasks are known Reassignment of task is not possible
- 5. Inter-task Communication Cost (cij) t1 t2 T3 t4 t1 0 35 3 8 t2 35 0 6 4 t3 3 6 0 23 t4 8 4 23 0 Arbitrary Assignment Task Node t1 n2 t2 n2 t3 n2 t4 n3 Optimal Assignment Task Node t1 n2 t2 n2 t3 n1 t4 n1 Execution Costs(xab) Task Nodes n1 n2 n3 t1 31 4 14 t2 1 5 6 t3 2 4 ∞ t4 3 28 10 An infinite cost for a particular task against a particular node indicate that the task cannot be executed on that node due to the task's requirement of specific resources that are not available on that node. Cost of arbitrary assignment (Total execution and communication cost) = x12 + x22 + x32 + x43+ c14 + c24 + c34 = 4 + 5 + 10 + 8 + 4 + 23 = 58 Cost of optimal assignment (total execution and communication cost) = x12 + x22 + x31 + x41+ c13 + c23 + c14 + c24 = 4 + 5 + 2 + 3 + 3 + 6 + 8 + 4 = 38
- 6. Processes submitted by the user are distributed among the nodes of the system so as to equalize the workload among the nodes. The goal of load-balancing algorithm is to maintain that each node is neither overloaded nor idle in order to obtain maximum performance of the system Load-balancing Approach is of two types:- Static Load-balancing Dynamic Load-balancing
- 7. – The processes are assign to each nodes at the compile time according to the performance of the nodes. – Once the processes are assigned, no change or reassignment is possible in the run time. – Number of jobs in each node is fixed and the algorithm do not collect any information about the nodes. – The assignment of jobs is done to the processing nodes on the basis of the following factors: incoming time, extent of resource needed, mean execution time and inter-process communication. – Since these factors should be measured before the assignment , this is why is called probabilistic algorithm
- 8. – In dynamic load-balancing algorithm the assignment is done at the run time. – Jobs are reassigned at the run time depending upon the situation that is the load will be transferred from heavily loaded nodes to the lightly loaded nodes. – In dynamic load balancing no decision is taken until the process gets execution. – This strategy collect the information about the system state and about the job information. – As more information is collected by an algorithm in a short time, potentially the algorithm can make better decision. – Dynamic approach is mostly considered in heterogeneous system because it consists of nodes with different speeds, different memory sizes and different communication link speed
- 9. – Load balancing approaches attempt to equalize the workload on all the nodes of a system by gathering state information – Load sharing algorithms do not attempt to balance the average workload on all nodes, they only ensure that no node is idle or heavily loaded – Policies for load sharing approach are the same as load balancing polices, they include load estimation policy, process transfer policy, location policy and state information exchange, they differ in location policy
- 10. ● No prior knowledge about processes User does not want to specify information about characteristic and requirements ● Dynamic in nature Decision should be made base on changing load of nodes and not fixed static policy ● Quick decision-making capability Algorithm must make quick decision about the assignment of task to nodes of system
- 11. ● Balanced system performance and scheduling overhead Great amount of information gives more intelligent decision, but increases overhead ● Stability ✔ Unstable when all processes are migrating without accomplishing any useful work ✔ It occurs when the nodes turn from lightly- loaded to heavily-loaded state and vice verse
- 12. ● Scalability A scheduling algorithm should be capable of handling small as well as large networks ● Fault tolerance Should be capable of working after the crash of one or more nodes of the system ● Fairness of service More users initiating equivalent processes expect to receive the same quality of service
- 13. To make accessibility of data in distributed environment and mange the resources over the autonomous network system, we have to maintain such a system which is efficient, flexible, portable and secure.